My understanding of smteees!™ was that doing them at the base layer would slow the chain, too much information to be tracked in memory, or some such.
This is why we are patiently waiting for magi to launch.
Which I keep seeing promises of announcements, but nothing I can use.
You are viewing a single comment's thread from:
Hmm, that's interesting. Deterministic SMT processing with strict per-block caps implemented would most likely prevent chain slowdown. They're not storing anymore information than operations currently which might might contain custom JSON. I'll look further into things.
Bt has done wonders with the memory requirements, much lower than when smteees!™ was first proposed.
From what I could see, it shouldn't be a problem tbh. I don't have complete contextual awareness of the codebase and prior decisions. But I think it should be fine and it's honestly a needed change for Hive to get them.
If you can get them done, I'm all for it.
We need as many attractive options as we can get, at this point.